Propulsion motor nondestructive testing method and device based on acoustic signals

Abstract

The invention discloses a propulsion motor nondestructive testing method, device and equipment based on an acoustic signal and a medium. The method comprises the following steps: acquiring a detection acoustic signal during operation of a propulsion motor through a sensor; converting the detection acoustic signal into a corresponding detection energy density curve; according to the deviation condition between the detection energy density curve and a reference curve in a normal sound library and between the detection energy density curve and a fault curve in a fault sound library, whether the propulsion motor has a fault or not is determined, and fault classification is determined when the fault exists; and updating the normal sound bank or the fault sound bank according to the detection energy density curve, thereby realizing early accurate identification of internal and external faults of the propulsion motor, and effectively reducing fault maintenance cost and accident risk.

Description

Technical Field

[0001] This application relates to the field of marine propulsion motor testing technology, specifically to a method and apparatus for non-destructive testing of propulsion motors based on acoustic signals. Background Technology

[0002] In the field of condition monitoring for critical equipment such as marine propulsion motors, existing fault monitoring technologies have significant limitations. Traditional solutions often rely on contact-based methods such as vibration sensors or temperature detection, which are ill-suited to the complex environmental conditions at sea, including high temperatures, high humidity, and salt spray. Contact sensors need to be directly mounted on the equipment surface, and long-term exposure to harsh sea conditions can easily lead to equipment corrosion, signal drift, or decreased measurement accuracy. Furthermore, they cannot achieve non-contact, non-destructive early fault identification. For example, faults such as bearing wear, insufficient lubrication, and loose engine mounts often manifest as weak anomalies in their early stages. Traditional methods, due to the limitations of contact deployment, cannot effectively capture such signals, resulting in delayed fault warnings and increasing the risk of equipment damage.

[0003] Furthermore, existing technologies are mostly limited to the independent analysis of data from a single device, lacking intelligent monitoring methods based on multi-device collaboration. Ship propulsion systems typically contain multiple motors of the same model, but traditional methods do not fully integrate horizontal comparisons of operational data from multiple devices, making it difficult to distinguish between normal operating fluctuations and genuine fault characteristics, resulting in a persistently high false alarm rate.

[0004] Therefore, there is an urgent need for a technology that can efficiently and accurately identify propulsion motor faults and adapt to complex marine environments. Summary of the Invention

[0005] This application provides a method, apparatus, equipment, and medium for non-destructive testing of propulsion motors based on acoustic signals, which can solve the technical problem that existing technologies cannot accurately and efficiently identify propulsion motor faults.

[0006] In a first aspect, embodiments of this application provide a non-destructive testing method for a propulsion motor based on acoustic signals, the method comprising: Acoustic signals detected during the operation of the propulsion motor are collected by sensors; The detected acoustic signal is converted into a corresponding detection energy density curve; Based on the deviation of the detected energy density curve from the baseline curve in the normal sound library and the fault curve in the fault sound library, it is determined whether the propulsion motor has a fault, and if a fault exists, the fault classification is determined. The normal sound library or the faulty sound library is updated based on the detected energy density curve.

[0007] In conjunction with the first aspect, in one embodiment, before determining whether the propulsion motor is faulty based on the deviation of the detected energy density curve from the reference curve in the normal sound library and the fault curve in the fault sound library, the method further includes: Collect normal acoustic signals when the propulsion motor is running normally, and fault acoustic signals when the propulsion motor is running under various types of faults; The normal acoustic signal is converted into a normal energy density curve and the fault acoustic signal is converted into a fault energy density curve by using Fourier analysis, wavelet analysis and Lie group representation algorithm. The baseline curve is obtained by clustering the normal energy density curve using the fuzzy C-means clustering algorithm, and the fault curve is obtained by clustering the fault energy density curve. The baseline curve is stored in the normal sound library, and the fault curve is stored in the fault sound library.

[0008] In conjunction with the first aspect, in one embodiment, storing the fault curve in the fault sound library further includes: Based on the acoustic characteristics of the fault acoustic signal, a mathematical model of the fault sound is established; Establish sound mathematical models for periodic faults (PS{T,f,v,s}), persistent faults (LS{f1,f2,T,t0,v,d}), and sudden faults (SS{f1,f2,v,d,T,t0,v0,d0}), and store them in the fault sound library. Where T is the period, f is the zeroth frequency, f1 is the first frequency, f2 is the second frequency, v is the first volume, v0 is the zeroth volume, s is the period stability, t0 is the start time, d is the first density, and d0 is the second density.

[0009] In conjunction with the first aspect, in one implementation, based on the deviation of the detected energy density curve from the reference curve in the normal sound library and the fault curve in the fault sound library, it is determined whether the propulsion motor has a fault, and if a fault exists, a fault classification is determined, including: Determine the deviation values ​​of the detection energy density curve from the reference curve and the fault curve, respectively; Determine whether the minimum deviation value is greater than the preset deviation threshold; When the minimum deviation value is less than or equal to the deviation threshold, if the minimum deviation value is the deviation value between the detected energy density curve and the reference curve, then it is determined that the propulsion motor is operating normally; if the minimum deviation value is the deviation value between the detected energy density curve and the fault curve, then it is determined that the propulsion motor has a fault, and the fault type is determined based on the period, frequency, volume, density and stability periodicity of the detected acoustic signal, combined with the fault sound mathematical model. When the minimum deviation value is greater than the deviation threshold and exceeds a preset number of times, a prompt is made to perform manual fault detection.

[0010] In conjunction with the first aspect, in one implementation, updating the normal sound library or the faulty sound library based on the detected energy density curve includes: After determining that the propulsion motor is operating normally based on the detected energy density curve, the detected energy density curve is used as the reference curve and stored in the normal sound library; After determining that the propulsion motor has a fault based on the detected energy density curve, the detected energy density curve is stored as the fault curve in the fault sound library.

[0011] In conjunction with the first aspect, in one implementation, the method further includes: Calculate the overall deviation of the energy and density of the detection acoustic signals of two propulsion motors of the same model in the same frequency band; If the overall deviation is greater than the preset deviation threshold, and the volume density or periodic sound of the acoustic signal of either of the two propulsion motors of the same model is greater than the corresponding judgment threshold, then a propulsion motor fault warning is triggered.

[0012] In conjunction with the first aspect, in one embodiment, the sensor is deployed in the drive end bearing area, non-drive end bearing area, main push motor housing area, main push motor foot area, and side push motor flange area of ​​the propulsion motor by magnetic attraction or adhesive.

[0013] Secondly, embodiments of this application provide a non-destructive testing device for a propulsion motor based on acoustic signals, the non-destructive testing device for the propulsion motor based on acoustic signals comprising: The acquisition module is used to acquire the acoustic signals detected during the operation of the propulsion motor through sensors; A conversion module is used to convert the detected acoustic signal into a corresponding detection energy density curve; The determination module is used to determine whether the propulsion motor has a fault based on the deviation of the detected energy density curve from the reference curve in the normal sound library and the fault curve in the fault sound library, and to determine the fault classification when a fault exists. An update module is used to update the normal sound library or the faulty sound library based on the detected energy density curve.

[0014] Thirdly, embodiments of this application provide a propulsion motor non-destructive testing device based on acoustic signals. The propulsion motor non-destructive testing device based on acoustic signals includes a processor, a memory, and a propulsion motor non-destructive testing program based on acoustic signals stored in the memory and executable by the processor. When the propulsion motor non-destructive testing program based on acoustic signals is executed by the processor, it implements the steps of the propulsion motor non-destructive testing method based on acoustic signals as described in any of the preceding claims.

[0015] Fourthly, embodiments of this application provide a computer-readable storage medium storing a non-destructive testing program for a propulsion motor based on acoustic signals, wherein when the non-destructive testing program for a propulsion motor based on acoustic signals is executed by a processor, it implements the steps of the non-destructive testing method for a propulsion motor based on acoustic signals as described in any of the preceding claims.

[0016] The beneficial effects of the technical solutions provided in this application include: Acoustic signals from the propulsion motor during operation are collected by sensors; these signals are converted into corresponding energy density curves; based on the deviations of the energy density curves from a baseline curve in a normal sound library and a fault curve in a fault sound library, the presence of a fault in the propulsion motor is determined, and if a fault is found, a fault classification is determined; the normal sound library or the fault sound library is updated based on the energy density curves. Multi-device comparison and baseline library matching based on acoustic signals improve diagnostic accuracy, reduce false alarms, support real-time early warning and fault classification, and achieve early and accurate identification of internal and external faults in the propulsion motor, effectively reducing maintenance costs and accident risks. Attached Figure Description

[0017] Figure 1 This is a schematic flowchart of an embodiment of the non-destructive testing method for propulsion motors based on acoustic signals according to this application; Figure 2 This is a schematic diagram of the functional modules of an embodiment of the non-destructive testing method and apparatus for propulsion motors based on acoustic signals according to this application; Figure 3 This is a schematic diagram of the hardware structure of the non-destructive testing equipment for propulsion motors based on acoustic signals involved in the embodiments of this application. Detailed Implementation

[0018] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present application.

[0019] To make the objectives, technical solutions, and advantages of this application clearer, the embodiments of this application will be described in further detail below with reference to the accompanying drawings.

[0020] In a first aspect, embodiments of this application provide a non-destructive testing method for propulsion motors based on acoustic signals.

[0021] In one embodiment, reference is made to Figure 1 , Figure 1 This is a flowchart illustrating the first embodiment of the non-destructive testing method for propulsion motors based on acoustic signals according to this application. Figure 1 As shown, the non-destructive testing method for propulsion motors based on acoustic signals includes: Step S101: Acquire the detection acoustic signal of the propulsion motor during operation through the sensor.

[0022] Step S102: Convert the detected acoustic signal into a corresponding detection energy density curve.

[0023] Step S103: Based on the deviation between the detected energy density curve and the reference curve in the normal sound library and the fault curve in the fault sound library, determine whether the propulsion motor has a fault, and if a fault exists, determine the fault classification.

[0024] Step S104: Update the normal sound library or the faulty sound library according to the detected energy density curve.

[0025] This embodiment improves the accuracy of propulsion motor fault diagnosis and reduces false alarms by matching the detected acoustic signals during propulsion motor operation with curves in the normal sound library and fault sound library. It supports real-time early warning and fault classification, and realizes early and accurate identification of internal and external faults of propulsion motor, which can effectively reduce fault maintenance costs and accident risks.

[0026] It is worth noting that before conducting non-destructive testing of propulsion motors based on acoustic signals, it is necessary to deploy sensors to collect acoustic signals during propulsion motor operation and establish normal sound libraries and fault sound libraries.

[0027] In one embodiment, the acoustic sensor deployment includes: deploying the acoustic sensor in the drive end bearing area, non-drive end bearing area, main push motor housing area, main push motor foot area, and side push motor flange area of ​​the propulsion motor by magnetic attraction or adhesive.

[0028] On the one hand, regarding the best installation of sensors: The installation location of the sensors is determined based on a deep understanding of the failure mechanism of the propulsion motor and the characteristics of sound propagation, aiming to capture the acoustic features that best reflect the health status of the propulsion motor. (1) Strategic deployment: Acoustic sensors (sensor earpieces) can be deployed in key locations where the propulsion motor is prone to failure. For example: Drive end bearing / non-drive end bearing area: mainly used to detect bearing failure and insufficient lubrication. During long-term sailing, the bearings bear continuous loads and friction, as well as oxidation and corrosion caused by high temperature, high humidity, and salt air at sea, making them prone to wear. Deploying listening points in this area can detect abnormalities in the early stage of failure. Main propulsion motor housing area: mainly used to detect problems such as broken rotor bars and imbalance caused by the drive. Special marine environments such as high temperature, high humidity, and high vibration may damage the rotor, leading to broken bars and imbalance. Main propulsion motor foot area: mainly used to detect loose mounting and wear problems of the foot. Vibration, improper installation, etc. may cause the foot fastening bolts to loosen, which in turn causes vibration and noise, and aggravates the wear of other components. Side thrust motor flange area: used to detect bearing problems and motor operating status (such as abnormal speed). (2) Number and illustration of deployment points: In this embodiment, a total of 8 sound data acquisition points are set, including 1 drive end bearing, 1 non-drive end bearing, 4 machine feet, and 2 housings. (3) Installation method and environmental adaptability: The sensor earpiece adopts magnetic installation. Different earpiece sensors can be customized according to the on-site equipment installation environment to ensure its matching with the specific installation area. In noisy environments, magnetic or adhesive adsorption installation can be used to reduce the impact of environmental wind speed and industrial site signal interference on data reliability. In addition, the sensor shell is designed with corrosion-resistant materials to adapt to the harsh marine environment. (4) Dynamic adjustment: During the sensor installation verification test, the test personnel will make fine adjustments to the test points according to the test results to better demonstrate the fault detection effect. The selection of the installation position has a certain degree of flexibility and experience optimization space.

[0029] On the other hand, regarding the sampling frequency of the sensor: the system's choice of sampling frequency reflects the need for full-band sound signal capture and high-precision analysis to cope with the complexity of industrial noise. (1) High sampling frequency and depth: The stethoscope has an ultra-high frequency acquisition frequency, supports the selection of 16 / 24 / 32-bit sampling depth and 0~640kHz sampling frequency to adapt to the acquisition needs of various sound signals, and more accurately restore the original sound signal to obtain more accurate sound data. (2) Full-band coverage: The system can achieve accurate quantization and analysis of sound signals from infrasound to ultrasound. It includes the ability to comprehensively analyze ultra-high frequency electromagnetic noise that is inaudible to the human ear and ultra-high frequency metal impact sound. High sampling frequency is a necessary condition for capturing these high-frequency and ultra-high frequency signals. According to the Nyquist sampling theorem, the sampling frequency should be set to at least twice the highest frequency of the measured signal. High frequency acquisition frequency and depth ensure the reliability and comprehensiveness of the acquired data, providing data support for subsequent real-time calculation and processing.

[0030] After the acoustic sensors are deployed, acoustic signals can be collected using the deployed sensors. Based on the collected acoustic signals, normal sound libraries and fault sound libraries can be built. Specific steps include: The system collects normal acoustic signals during normal operation of the propulsion motor, as well as fault acoustic signals during operation under various types of faults. Using Fourier analysis, wavelet analysis, and Lie group representation algorithms, the normal acoustic signals are converted into normal energy density curves, and the fault acoustic signals are converted into fault energy density curves. A fuzzy C-means clustering algorithm is used to cluster the normal energy density curves to obtain the baseline curve, and the fault energy density curves are clustered to obtain the fault curve. The baseline curve is stored in the normal sound library, and the fault curve is stored in the fault sound library.

[0031] As an example, the non-destructive testing of propulsion motors based on acoustic signals in this application includes a collection period, an early warning period, and a maintenance period, with the period for generating normal sound libraries and fault sound libraries being the collection period.

[0032] The establishment of a normal sound database during the collection period includes: continuously collecting acoustic signals of the propulsion motor during normal operation at second intervals (e.g., 3 seconds) within a preset first duration (e.g., 48 hours), and recording these as normal acoustic signals. In the verification experiment, nine typical operating conditions of the propulsion motor—50%, 60%, and 100% speed, no-load, 50% load, and 80% load—are selected, and sound data is collected 10 times from the normally operating propulsion motor as normal acoustic signals. Frequency and time domain information of the normal acoustic signals is obtained by improving Fourier analysis, wavelet analysis, and Lie group representation algorithms, thereby converting the normal acoustic signals into normal energy density curves. Then, cluster analysis (fuzzy C-means cluster analysis in this embodiment) is used to classify these curves, forming multiple benchmark curves representing the normal operating state of the propulsion motor. These benchmark curves are then stored in the normal sound database, generating the normal sound database.

[0033] Establishing a fault sound library involves simulating propulsion motor faults using a fault simulation system. This system includes a non-destructive testing (NDT) bench for propulsion motors, capable of simulating three typical faults: insufficient installation precision, loose fasteners, and bearing wear. For example, artificially controlling the radial displacement of the propulsion motor simulates rotor misalignment due to insufficient installation precision; artificially loosening the screws on the propulsion motor's feet simulates loose fasteners; and manufacturing a faulty motor and grooving the inner ring of the bearing simulates bearing wear. The sound data collected under these simulated fault conditions serves as the acoustic signals of the propulsion motor faults. This fault sound data can be directly used to verify the system's fault identification capabilities.

[0034] For fault acoustic signals, frequency and time domain information can be obtained by improving Fourier analysis, wavelet analysis, and Lie group representation algorithms, thereby converting the fault acoustic signals into fault energy density curves. Then, fuzzy C-means clustering analysis is used to classify the fault energy density curves, forming multiple fault curves representing the fault state of the propulsion motor. These fault curves are then stored in a fault sound database to obtain a fault sound library.

[0035] Explanatoryly, regarding the adaptability of the signal processing algorithm in this embodiment, the required algorithm must be able to process complex, variable, and noisy industrial sound signals, as well as have a mechanism for self-learning and dynamic correction under different operating conditions.

[0036] In this embodiment, a mathematical analysis method based on improved Fourier analysis, wavelet analysis, and Lie group representation is selected to generate the energy density curve of the acoustic signal. This combination can accurately acquire the frequency and time domain information of the sound signal, realizing the statistical analysis of all sound information. Unlike traditional Fourier analysis, which is not suitable for complex, non-periodic, sudden, and rapidly decaying actual industrial noise signals, this improved algorithm can construct a fully quantifiable statistical acoustic model, enabling high-precision quantitative analysis of periodic, continuous, and sudden sounds.

[0037] Fuzzy C-means clustering (FCM) is the core algorithm for classifying sound data in this embodiment. During the collection period, FCM clusters and classifies the sound data of the motor during normal operation, forming multiple baseline curves representing the normal operating state, which are stored in a normal sound library. FCM can implicitly and automatically establish baselines for the normal state under different operating conditions (such as load changes, speed changes, and even indirectly reflecting the impact of sea state fluctuations on motor operation) based on sound characteristics, without the need to explicitly input these operating condition parameters. FCM can effectively handle noise and ambiguity, allowing data points to belong to multiple clusters simultaneously with their membership degree. The clustering results can overlap, and it can handle noisy data well because the membership degree of noise points is lower, reducing the impact on the clustering results. This is crucial for noisy and highly uncertain marine environments.

[0038] In a preferred embodiment, when storing the fault curve in the fault sound library, the method further includes: establishing a fault sound mathematical model based on the sound characteristics of the fault acoustic signal, and storing it in the fault sound library. Establishing the fault sound mathematical model includes: establishing a sound mathematical model PS{T,f,v,s} for periodic faults, a sound mathematical model LS{f1,f2,T,t0,v,d} for persistent faults, and a sound mathematical model SS{f1,f2,v,d,T,t0,v0,d0} for sudden faults. Wherein, T is the period, f is the zeroth frequency, f1 is the first frequency, f2 is the second frequency, v is the first volume, v0 is the zeroth volume, s is the periodic stability, t0 is the start time, d is the first density, and d0 is the second density.

[0039] It is worth noting that these models and their parameters can quantify and analyze different types of sounds, and combine this with information such as the transmission cycle of equipment components to determine the cause of failure or assess health status. This mathematical model-based quantification decomposition does not require a large amount of abnormal sample data or the establishment of learning models, which has significant advantages in complex industrial (especially marine) environments where it is difficult to obtain comprehensive failure samples.

[0040] After establishing the normal sound library and the fault sound library, the early warning period can begin. During the early warning period, detection acoustic signals of the propulsion motor during operation can be collected using pre-deployed sensors. The collected detection acoustic signals can be converted into detection energy density curves using Fourier analysis, wavelet analysis, and Lie group representation algorithms.

[0041] In one embodiment, during the warning period, based on the deviation of the detected energy density curve from the reference curve in the normal sound library and the fault curve in the fault sound library, it is determined whether the propulsion motor is faulty, and if a fault is found, a fault classification is determined, including: determining the deviation values ​​of the detected energy density curve from the reference curve and the fault curve respectively; determining whether the minimum deviation value is greater than a preset deviation threshold; when the minimum deviation value is less than or equal to the deviation threshold, if the minimum deviation value is the deviation value of the detected energy density curve from the reference curve, it is determined that the propulsion motor is operating normally; if the minimum deviation value is the deviation value of the detected energy density curve from the fault curve, it is determined that the propulsion motor is faulty, and the fault type is determined based on the period, frequency, volume, density, and stability periodicity of the detected acoustic signal, combined with the fault sound mathematical model; when the minimum deviation value is greater than the deviation threshold and exceeds a preset number of times, a prompt is made to perform manual fault detection.

[0042] As an example, after entering the warning period, new acoustic signals of the propulsion motor will be collected by the deployed sensors and recorded as detection acoustic signals. After converting the detection acoustic signals into the corresponding detection energy density curves, the deviation value P of the detection energy density curves is calculated with respect to the baseline curves in the normal sound library and the fault curves in the fault sound library, and the minimum deviation value Pmin among all deviation values ​​P is found. It is determined whether Pmin is greater than the preset sound matching deviation value threshold P0. If Pmin is greater than P0, the system records the number of mismatches of the detection energy density curves m = m + 1. If the cumulative number of mismatches exceeds M(3), an early warning is triggered, and manual intervention is required for fault detection and judgment. This is to filter out signal mismatches caused by gross errors, thereby reducing the workload of manual intervention. If Pmin is less than or equal to P0 (and the number of mismatches m is less than 3), it means that the detection energy density curve matches a certain baseline curve in the two databases. If the matching sound comes from the normal sound library, the system determines that the equipment is in normal operating condition. If the matching sound comes from the fault sound library, the system triggers an alarm and informs the cause of the fault.

[0043] It is worth noting that the principle behind determining the fault type based on the period, frequency, volume, density, and periodicity of the detected acoustic signal, combined with the fault sound mathematical model, is as follows: the fault sound mathematical model represents the mapping relationship between the four major characteristics of the acoustic signal and the fault type. The monitoring platform can accurately identify abnormal noise and classify it into four major types, and then associate it with different fault types. This mapping relationship is mainly achieved through the quantitative analysis of key parameters such as the energy, density, period, frequency, and periodicity stability of the sound signal.

[0044] (1) New periodic noise appears: Characteristic description: During normal operation of the equipment, periodic sounds are usually stable. If new periodic noise appears, it indicates an anomaly. Mapped fault type: This type of noise may indicate problems such as bearing misalignment or loose screws. It is related to the mathematical model of periodic sound PS{T, f, v, s}. The cause of the fault is determined by analyzing the period (T), frequency (f), volume (v), and periodic stability (s), combined with the transmission cycle of the equipment components. For example, loose machine feet assembly can cause the machine feet to loosen, resulting in vibration and noise. Frequency range / density threshold: Based on the quantitative analysis of T, f, v, s, especially the abnormal occurrence of f (frequency) and s (periodic stability).

[0045] (2) Increase in existing periodic sound. Characteristic description: During normal operation of the equipment, the existing periodic sound gradually or suddenly increases. Mapped fault type: This change usually points to problems such as reduced lubrication and gear wear. It is also related to the mathematical model of periodic sound PS{T, f, v, s}, with the focus on the increase of v (volume). Insufficient lubrication is a specific problem that the system can detect and remind the user to replenish it when lubrication takes effect. Wear faults are manifested in the energy and density curves as an increase in both energy and density in the mid-to-high frequency range. Frequency range / density threshold: For problems such as bearing wear and reduced lubrication, energy and density may increase in the low, mid, and high frequency ranges, especially the density and energy in the high frequency range.

[0046] (3) New persistent noise at a specific frequency appears. Characteristic description: The equipment emits persistent noise at a specific frequency without a periodic pattern. Mapped fault type: This type of noise may be an indication of faults such as longitudinal cracking of the bearing or breakage of steel balls. It is related to the mathematical model of persistent sound LS{f1, f2, T, t0, v, d}, and the machine health status is assessed by focusing on the changes in sound (volume v and density d) in a specific sound frequency range (f1-f2). Frequency range / density threshold: The sound characteristics of cracking faults are a decrease in the energy and density of sound in the low, mid, and high frequency ranges, with a particularly significant decrease in the high frequency range. That is, the decrease in energy and density within a specific frequency range serves as an indicator of the appearance of new persistent noise.

[0047] (4) Increase in existing continuous noise at a specific frequency. Characteristic description: The existing continuous noise at a specific frequency increases continuously or suddenly. Mapped fault type: This usually indicates severe wear of the equipment. It is also related to the mathematical model of continuous sound LS{f1, f2, T, t0, v, d}, mainly focusing on the increase in volume (v) and density (d) within the existing frequency band. Frequency range / density threshold: For wear faults, density and energy increase to varying degrees in the low, mid, and high frequency bands. For eccentric faults, the sound energy increases while the density decreases in the high frequency band. These specific frequency responses and energy / density changes can serve as the basis for judging the increase in existing continuous noise.

[0048] Furthermore, in an optional embodiment, updating the normal sound library or the fault sound library based on the detected energy density curve includes: after determining that the propulsion motor is operating normally based on the detected energy density curve, storing the detected energy density curve as the reference curve in the normal sound library; and after determining that the propulsion motor has a fault based on the detected energy density curve, storing the detected energy density curve as the fault curve in the fault sound library.

[0049] As an example, if a manually confirmed energy density curve does not match either the normal sound library or the fault sound library, but is actually new normal operating data for the propulsion motor (e.g., the normal sound of the equipment under a new load or sea state, and the sound pattern of this condition is not yet fully covered by the "normal sound library"), then the detected energy density curve will be entered into the normal sound library to continuously optimize the model, thereby expanding the coverage of the statistical acoustic model's definition of "normal" and enabling it to adapt to a wider range of operating conditions. If the manual confirmation result indicates that the propulsion motor is faulty, it can be considered that the detected acoustic signal corresponds to a new fault, triggering an alarm and informing the cause of the fault. Simultaneously, the corresponding detected energy density curve is entered into the fault sound library, and the cause of the fault is marked. In this way, the system collaborates with the user to build a fault sound library, visualizing abnormal sounds and accurately matching them to faulty components. The fault sound library is built through alarms for deviations in frequency band volume and density, as well as specific equipment faults reported by maintenance personnel.

[0050] During the Maintenance Period, based on the analysis results from the Early Warning Period, the system will provide functions such as data presentation, degradation trend analysis, intelligent alarm system, and maintenance suggestions to support preventive maintenance and fault handling of equipment.

[0051] In an optional embodiment, the internal parameters of the FCM algorithm (e.g., cluster centers, membership functions) are adjusted based on new data, enabling the model to better classify and distinguish between normal and abnormal data. The inherent advantages of the FCM algorithm allow it to handle noisy data and highly fuzzy datasets effectively. In complex marine noise environments (such as seawater flow, mechanical operation, wind noise, vibration, etc.), sound signals often exhibit high complexity and uncertainty. The fuzzy nature of FCM allows for more flexible processing of these signals, allowing data points to belong to multiple clusters with a certain membership degree, thereby reducing the impact of noise on clustering results. Through the aforementioned manual intervention mechanisms of the "collection period" and "early warning period," normal sound changes caused by various sea state fluctuations encountered during actual system operation can be included in the "normal sound database" after manual confirmation, enabling the statistical acoustic model to continuously learn and adapt to the normal sound characteristics under different sea conditions.

[0052] Fuzzy C-means clustering (FCM) is a clustering algorithm based on fuzzy logic. It allows data points to belong to multiple clusters simultaneously with a certain membership degree, rather than strictly classifying them into a single category. This characteristic makes it well-suited for datasets with high fuzziness, which is crucial for complex industrial sound data that may contain transitional states. Handling overlap and noise: FCM allows for overlap in clustering results and can handle noisy data well because noise points have lower membership degrees, reducing their impact on the clustering results. This is a significant advantage for sound signals collected in noisy industrial environments. Optimization objective: FCM clusters by minimizing an objective function (JF), which measures the weighted distance from each sample point to each cluster center, introducing a fuzzy weighting exponent m. The algorithm iteratively optimizes by calculating the fuzzy partition matrix and updating the cluster center matrix until the objective function reaches its minimum. Parameter sensitivity: It is worth noting that the FCM algorithm is relatively sensitive to parameter selection, and parameter selection has a certain degree of subjectivity. Therefore, when using FCM, it is necessary to determine an appropriate matching deviation threshold P0 for the audio data.

[0053] In an optional embodiment, the non-destructive testing method for propulsion motors based on acoustic signals further includes: Calculate the overall deviation of the energy and density of the detection acoustic signals of two propulsion motors of the same model in the same frequency band; if the overall deviation is greater than a preset deviation threshold, and the volume density or periodic sound of the acoustic signal of either of the two propulsion motors of the same model is greater than the corresponding judgment threshold, then a propulsion motor fault warning is triggered.

[0054] The core of exemplary industrial acoustic spectrum technology lies in its simultaneous consistency and discriminative power, allowing for the definition of deviations. This embodiment utilizes a difference quantification method to integrate the LP norm of the energy and density curves of acoustic signals from different propulsion motors of the same model, achieving fault detection. The system performs a lateral comparative analysis of two sets of sound data from two propulsion motors of the same type but operating independently at the same time point, calculating the difference in energy (dB) and density (den) in the same frequency band for each pair of data, denoted as e. Subsequently, these single-frequency band differences e are integrated using the LP norm to derive the overall deviation value Dist. This Dist value is the key indicator for quantifying significant differences.

[0055] The quantification standard is characterized by customization and adjustability; some parameters in the overall deviation value Dist formula can be adjusted according to actual conditions. For example, ω (Omega) is used to control the importance weight of different frequency bands. In specific application scenarios, the differences in some frequency bands may be more critical than those in others. The system allows assigning higher weights to these frequency bands, making them play a more important role in the overall deviation calculation. k, t, and b are used to adjust the relative proportions of energy and density. When evaluating differences, the system can, according to actual needs, focus more on changes in energy, more on changes in density, or both, to adapt to the different impacts of different types of fault modes on sound characteristics. This design philosophy reflects the system's flexibility and adaptability under different industrial noise and equipment operating conditions.

[0056] Next, significant differences can be used to determine faults in the propulsion motors and determine system responses. In a horizontal comparison, when the Dist value exceeds a preset deviation threshold, the system will consider two propulsion motors of the same type but operating independently to have significant differences. When sound data (whether volume density or periodic sound) reaches a "certain threshold" or a "set threshold," an alarm or warning will be triggered.

[0057] In an optional embodiment, machine learning technology is deeply integrated and plays a crucial role in the "fault diagnosis and alarm early warning mechanism" of this application. The system first extracts key feature values ​​from the collected acoustic signals. These feature values, after being filtered and processed, are input into a pre-built machine learning model for training. Using a large amount of historical fault data and normal operation data as training samples, the model can learn and identify the unique characteristics of different types of fault modes, thereby achieving accurate classification and identification of faults. This process not only improves the accuracy of fault diagnosis but also quickly distinguishes between normal operation and various fault states, providing support for timely measures and effectively extending the service life of equipment, reducing equipment downtime and maintenance costs. By deeply integrating machine learning with big data technology, this application not only enhances the intelligence level of fault diagnosis but also realizes a shift from passive maintenance to proactive prevention in the maintenance model, providing strong protection for the safe and efficient operation of motor devices.

[0058] It is worth noting that in this embodiment, the acoustic sensor is responsible for collecting sound data, specifically for the data transmission format. The collected sound data is transmitted to an edge hardware box for "preprocessing." After preprocessing, the data is further processed by algorithms and compressed into a binary sound file containing acoustic feature values ​​before being uploaded to the server. This process ensures that the data is effectively processed and compressed before transmission, thereby improving the efficiency and security of data transmission. Regarding network communication fault tolerance mechanisms, this system has a network outage caching function. Even without a network connection, the sensor can still retain monitoring data in the background. When the network is restored, this temporarily stored data will be automatically uploaded to the server, effectively preventing data loss. This mechanism greatly enhances the system's reliability in unstable network environments.

[0059] Furthermore, this system supports multiple network transmission methods, including Wi-Fi, fiber optic, and 4G / 5G networks. This diverse range of communication options significantly enhances the system's network adaptability and robustness. In the face of a single network failure, the system can flexibly switch to other available networks, thereby greatly reducing the risk of data transmission interruptions due to network problems and ensuring the continuity and stability of data transmission.

[0060] This application provides a non-destructive testing method for propulsion motors based on acoustic signals. This method utilizes advanced acoustic signal processing technology to achieve early and accurate identification of internal and external faults in propulsion motors. Testing has verified that the method achieves a fault identification rate exceeding 95% and a response time of less than 16 seconds, significantly improving the efficiency and timeliness of fault detection. By promptly identifying potential faults, this method effectively reduces equipment maintenance costs and accident risks, ensuring the safe and stable operation of the propulsion motor. Furthermore, the method employs multi-device comparative analysis and benchmark library matching techniques, further improving diagnostic accuracy and significantly reducing false alarms. The system supports real-time early warning functions, issuing alarms promptly in the early stages of fault occurrence and classifying faults to help maintenance personnel quickly locate problems and take targeted measures. The method also possesses the ability to analyze sound signals across the entire frequency band, enabling non-contact monitoring and avoiding the wear and interference problems that may arise from traditional contact monitoring equipment. Simultaneously, the non-contact monitoring method reduces equipment deployment costs and maintenance difficulty. In terms of diagnostic efficiency, this method can shorten the diagnostic time to the millisecond level, greatly improving the real-time performance and accuracy of fault detection, providing strong technical support for the efficient operation and maintenance of propulsion motors.

[0061] Secondly, embodiments of this application also provide a non-destructive testing device for propulsion motors based on acoustic signals.

[0062] In one embodiment, reference is made to Figure 2 , Figure 2 This is a functional module diagram of an embodiment of the acoustic signal-based non-destructive testing device for propulsion motors according to this application. Figure 2 As shown, the non-destructive testing device for propulsion motors based on acoustic signals includes: The acquisition module is used to acquire the acoustic signals detected during the operation of the propulsion motor through sensors; A conversion module is used to convert the detected acoustic signal into a corresponding detection energy density curve; The determination module is used to determine whether the propulsion motor has a fault based on the deviation of the detected energy density curve from the reference curve in the normal sound library and the fault curve in the fault sound library, and to determine the fault classification when a fault exists. An update module is used to update the normal sound library or the faulty sound library based on the detected energy density curve.

[0063] Furthermore, in one embodiment, the device is also used for: Collect normal acoustic signals when the propulsion motor is running normally, and fault acoustic signals when the propulsion motor is running under various types of faults; The normal acoustic signal is converted into a normal energy density curve and the fault acoustic signal is converted into a fault energy density curve by using Fourier analysis, wavelet analysis and Lie group representation algorithm. The baseline curve is obtained by clustering the normal energy density curve using the fuzzy C-means clustering algorithm, and the fault curve is obtained by clustering the fault energy density curve. The baseline curve is stored in the normal sound library, and the fault curve is stored in the fault sound library.

[0064] Furthermore, in one embodiment, the device is also used for: Based on the acoustic characteristics of the fault acoustic signal, a mathematical model of the fault sound is established; Establish sound mathematical models for periodic faults (PS{T,f,v,s}), persistent faults (LS{f1,f2,T,t0,v,d}), and sudden faults (SS{f1,f2,v,d,T,t0,v0,d0}), and store them in the fault sound library. Where T is the period, f is the zeroth frequency, f1 is the first frequency, f2 is the second frequency, v is the first volume, v0 is the zeroth volume, s is the period stability, t0 is the start time, d is the first density, and d0 is the second density.

[0065] Furthermore, in one embodiment, the determining module is further configured to: Determine the deviation values ​​of the detection energy density curve from the reference curve and the fault curve, respectively; Determine whether the minimum deviation value is greater than the preset deviation threshold; When the minimum deviation value is less than or equal to the deviation threshold, if the minimum deviation value is the deviation value between the detected energy density curve and the reference curve, then it is determined that the propulsion motor is operating normally; if the minimum deviation value is the deviation value between the detected energy density curve and the fault curve, then it is determined that the propulsion motor has a fault, and the fault type is determined based on the period, frequency, volume, density and stability periodicity of the detected acoustic signal, combined with the fault sound mathematical model. When the minimum deviation value is greater than the deviation threshold and exceeds a preset number of times, a prompt is made to perform manual fault detection.

[0066] Furthermore, in one embodiment, the update module is also used to: After determining that the propulsion motor is operating normally based on the detected energy density curve, the detected energy density curve is used as the reference curve and stored in the normal sound library; After determining that the propulsion motor has a fault based on the detected energy density curve, the detected energy density curve is stored as the fault curve in the fault sound library.

[0067] Furthermore, in one embodiment, the device is also used for: Calculate the overall deviation of the energy and density of the detection acoustic signals of two propulsion motors of the same model in the same frequency band; If the overall deviation is greater than the preset deviation threshold, and the volume density or periodic sound of the acoustic signal of either of the two propulsion motors of the same model is greater than the corresponding judgment threshold, then a propulsion motor fault warning is triggered.

[0068] Furthermore, in one embodiment, the sensor is deployed in the drive end bearing area, non-drive end bearing area, main push motor housing area, main push motor foot area, and side push motor flange area of ​​the propulsion motor by magnetic attraction or adhesive.

[0069] The functions of each module in the acoustic signal-based propulsion motor non-destructive testing device correspond to the steps in the acoustic signal-based propulsion motor non-destructive testing method embodiment, and their functions and implementation processes will not be described in detail here.

[0070] Thirdly, embodiments of this application provide a non-destructive testing device for a propulsion motor based on acoustic signals. The non-destructive testing device for a propulsion motor based on acoustic signals can be a computer (PC), server, or other device with data processing capabilities.

[0071] Reference Figure 3 , Figure 3 This is a schematic diagram of the hardware structure of the acoustic signal-based non-destructive testing equipment for propulsion motors involved in the embodiments of this application. In this embodiment, the acoustic signal-based non-destructive testing equipment for propulsion motors may include a processor, a memory, a communication interface, and a communication bus.

[0072] The communication bus can be of any type and is used to interconnect the processor, memory, and communication interface.

[0073] The communication interface includes input / output (I / O) interfaces, physical interfaces, and logical interfaces used for interconnecting internal components of the acoustic signal-based propulsion motor non-destructive testing equipment, as well as interfaces used for interconnecting the acoustic signal-based propulsion motor non-destructive testing equipment with other devices (such as other computing devices or user equipment). Physical interfaces can be Ethernet interfaces, fiber optic interfaces, ATM interfaces, etc.; user equipment can be displays, keyboards, etc.

[0074] Memory can be various types of storage media, such as random access memory (RAM), read-only memory (ROM), non-volatile RAM (NVRAM), flash memory, optical storage, hard disk, programmable ROM (PROM), erasable PROM (EPROM), electrically erasable PROM (EEPROM), etc.

[0075] The processor can be a general-purpose processor, which can call the acoustic signal-based propulsion motor non-destructive testing program stored in the memory and execute the acoustic signal-based propulsion motor non-destructive testing method provided in the embodiments of this application. For example, the general-purpose processor can be a central processing unit (CPU). The method executed when the acoustic signal-based propulsion motor non-destructive testing program is called can be referred to the various embodiments of the acoustic signal-based propulsion motor non-destructive testing method of this application, and will not be repeated here.

[0076] Those skilled in the art will understand that Figure 3 The hardware structure shown does not constitute a limitation of this application and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0077] Fourthly, embodiments of this application also provide a computer-readable storage medium.

[0078] The present application stores a non-destructive testing program for a propulsion motor based on acoustic signals on a computer-readable storage medium, wherein when the non-destructive testing program for a propulsion motor based on acoustic signals is executed by a processor, the steps of the non-destructive testing method for a propulsion motor based on acoustic signals as described above are implemented.

[0079] The method implemented when the acoustic signal-based propulsion motor non-destructive testing program is executed can be referred to in the various embodiments of the acoustic signal-based propulsion motor non-destructive testing method of this application, and will not be repeated here.

[0080] It should be noted that the sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0081] The terms "comprising" and "having," and any variations thereof, in the specification, claims, and accompanying drawings of this application are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to such process, method, product, or apparatus. The terms "first," "second," and "third," etc., are used to distinguish different objects, etc., and do not indicate a sequence, nor do they limit "first," "second," and "third" to different types.

[0082] In the description of the embodiments of this application, terms such as "exemplary," "for example," or "for instance" are used to indicate examples, illustrations, or explanations. Any embodiment or design described as "exemplary," "for example," or "for instance" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of terms such as "exemplary," "for example," or "for instance" is intended to present the relevant concepts in a concrete manner.

[0083] In the description of the embodiments of this application, unless otherwise stated, " / " means "or". For example, A / B can mean A or B. The "and / or" in the text is merely a description of the relationship between related objects, indicating that there can be three relationships. For example, A and / or B can mean: A exists alone, A and B exist simultaneously, and B exists alone. In addition, in the description of the embodiments of this application, "multiple" means two or more.

[0084] In some processes described in the embodiments of this application, multiple operations or steps are included in a specific order. However, it should be understood that these operations or steps may not be executed in the order they appear in the embodiments of this application, or they may be executed in parallel. The sequence number of the operation is only used to distinguish different operations, and the sequence number itself does not represent any execution order. In addition, these processes may include more or fewer operations, and these operations or steps may be executed sequentially or in parallel, and these operations or steps may be combined.

[0085] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes several instructions to cause a terminal device to execute the methods described in the various embodiments of this application.

[0086] The above are merely preferred embodiments of this application and do not limit the patent scope of this application. Any equivalent structural or procedural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

Claims

1. A non-destructive testing method for propulsion motors based on acoustic signals, characterized in that, The non-destructive testing method for propulsion motors based on acoustic signals includes: Acoustic signals detected during the operation of the propulsion motor are collected by sensors; The detected acoustic signal is converted into a corresponding detection energy density curve; Based on the deviation of the detected energy density curve from the baseline curve in the normal sound library and the fault curve in the fault sound library, it is determined whether the propulsion motor has a fault, and if a fault exists, the fault classification is determined. The normal sound library or the faulty sound library is updated based on the detected energy density curve.

2. The non-destructive testing method for propulsion motors based on acoustic signals as described in claim 1, characterized in that, Before determining whether the propulsion motor has a fault based on the deviation of the detected energy density curve from the reference curve in the normal sound library and the fault curve in the fault sound library, the method further includes: Collect normal acoustic signals when the propulsion motor is running normally, and fault acoustic signals when the propulsion motor is running under various types of faults; The normal acoustic signal is converted into a normal energy density curve and the fault acoustic signal is converted into a fault energy density curve by using Fourier analysis, wavelet analysis and Lie group representation algorithm. The baseline curve is obtained by clustering the normal energy density curve using the fuzzy C-means clustering algorithm, and the fault curve is obtained by clustering the fault energy density curve. The baseline curve is stored in the normal sound library, and the fault curve is stored in the fault sound library.

3. The non-destructive testing method for propulsion motors based on acoustic signals as described in claim 2, characterized in that, Storing the fault curve in the fault sound library also includes: Based on the acoustic characteristics of the fault acoustic signal, a mathematical model of the fault sound is established; Establish sound mathematical models for periodic faults (PS{T,f,v,s}), persistent faults (LS{f1,f2,T,t0,v,d}), and sudden faults (SS{f1,f2,v,d,T,t0,v0,d0}), and store them in the fault sound library. Where T is the period, f is the zeroth frequency, f1 is the first frequency, f2 is the second frequency, v is the first volume, v0 is the zeroth volume, s is the period stability, t0 is the start time, d is the first density, and d0 is the second density.

4. The non-destructive testing method for propulsion motors based on acoustic signals as described in claim 3, characterized in that, Based on the deviation of the detected energy density curve from the baseline curve in the normal sound library and the fault curve in the fault sound library, it is determined whether the propulsion motor has a fault, and if a fault exists, a fault classification is determined, including: Determine the deviation values ​​of the detection energy density curve from the reference curve and the fault curve, respectively; Determine whether the minimum deviation value is greater than the preset deviation threshold; When the minimum deviation value is less than or equal to the deviation threshold, if the minimum deviation value is the deviation value between the detected energy density curve and the reference curve, then it is determined that the propulsion motor is operating normally; if the minimum deviation value is the deviation value between the detected energy density curve and the fault curve, then it is determined that the propulsion motor has a fault, and the fault type is determined based on the period, frequency, volume, density and stability periodicity of the detected acoustic signal, combined with the fault sound mathematical model. When the minimum deviation value is greater than the deviation threshold and exceeds a preset number of times, a prompt is made to perform manual fault detection.

5. The non-destructive testing method for propulsion motors based on acoustic signals as described in claim 1, characterized in that, Updating the normal sound library or the faulty sound library based on the detected energy density curve includes: After determining that the propulsion motor is operating normally based on the detected energy density curve, the detected energy density curve is used as the reference curve and stored in the normal sound library; After determining that the propulsion motor has a fault based on the detected energy density curve, the detected energy density curve is stored as the fault curve in the fault sound library.

6. The non-destructive testing method for propulsion motors based on acoustic signals as described in claim 1, characterized in that, The method also includes: Calculate the overall deviation of the energy and density of the detection acoustic signals of two propulsion motors of the same model in the same frequency band; If the overall deviation is greater than the preset deviation threshold, and the volume density or periodic sound of the acoustic signal of either of the two propulsion motors of the same model is greater than the corresponding judgment threshold, then a propulsion motor fault warning is triggered.

7. The non-destructive testing method for propulsion motors based on acoustic signals as described in claim 1, characterized in that: The sensors are deployed in the drive end bearing area, non-drive end bearing area, main push motor housing area, main push motor foot area, and side push motor flange area of ​​the propulsion motor by magnetic attraction or adhesive.

8. A non-destructive testing device for a propulsion motor based on acoustic signals, characterized in that, The acoustic signal-based non-destructive testing device for propulsion motors includes: The acquisition module is used to acquire the acoustic signals detected during the operation of the propulsion motor through sensors; A conversion module is used to convert the detected acoustic signal into a corresponding detection energy density curve; The determination module is used to determine whether the propulsion motor has a fault based on the deviation of the detected energy density curve from the reference curve in the normal sound library and the fault curve in the fault sound library, and to determine the fault classification when a fault exists. An update module is used to update the normal sound library or the faulty sound library based on the detected energy density curve.

9. A non-destructive testing device for propulsion motors based on acoustic signals, characterized in that, The acoustic signal-based propulsion motor non-destructive testing equipment includes a processor, a memory, and an acoustic signal-based propulsion motor non-destructive testing program stored in the memory and executable by the processor. When the acoustic signal-based propulsion motor non-destructive testing program is executed by the processor, it implements the steps of the acoustic signal-based propulsion motor non-destructive testing method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a non-destructive testing program for a propulsion motor based on acoustic signals, wherein when the non-destructive testing program for a propulsion motor based on acoustic signals is executed by a processor, it implements the steps of the non-destructive testing method for a propulsion motor based on acoustic signals as described in any one of claims 1 to 7.

Images